J. Phys. Chem. B 2005, 109, 7399-7406
7399
Photocatalytic Reactivity of Surface Platinized TiO2: Substrate Specificity and the Effect of Pt Oxidation State Jaesang Lee and Wonyong Choi* School of EnVironmental Science and Engineering and Department of Chemistry, Pohang UniVersity of Science and Technology, Pohang 790-784, Korea ReceiVed: December 8, 2004; In Final Form: February 1, 2005
Surface platinized TiO2 (Pt/TiO2) has been frequently studied, but its photocatalytic reactivities reported in the literature are not consistent in some cases. To understand the discrepancies, the effects of Pt speciation on TiO2 on the photocatalytic degradation (PCD) of a few chlorinated organic compounds (trichloroethylene (TCE), perchloroethylene (PCE), dichloroacetate, etc.) were investigated with several Pt/TiO2 samples that were prepared differently. The oxidation state of Pt deposits was analyzed by X-ray photoelectron spectroscopy and was found to be the most important factor in determining the initial PCD rates of chlorinated organic compounds. TiO2 with oxidized Pt species (Ptox/TiO2) was less reactive than TiO2 with metallic Pt (Pt0/TiO2) for all substrates tested. In particular, Ptox/TiO2 strongly inhibited the PCD of TCE and PCE whereas it was more reactive than pure TiO2 for the PCD of other compounds. The photocurrents obtained with the Ptox/ TiO2 electrode were lower than those with the Pt0/TiO2 electrode, which was ascribed to the role of Ptox species as a recombination center. It is proposed that TCE adsorbed on Ptox chemically mediates the charge recombination through the redox cycle of TCE. The Pt effects in photocatalysis are highly substrate-specific and depend on the Pt-substrate interaction as well as the properties of Pt deposits.
Introduction TiO2 has been intensively investigated as a practical photocatalyst for environmental cleanup since the 1980s,1-6 and many methods have been developed to modify its catalytic performance. The surface platinization of TiO2 has been a popular photocatalyst modification technique, since Kraeutler and Bard7 first introduced it, because the platinized TiO2 (Pt/TiO2) exhibits enhanced activity for many photocatalytic reactions.7-10 The presence of Pt deposits on TiO2 is believed to retard fast chargepair recombination by serving as an electron sink (Schottky barrier electron trapping) and to facilitate the interfacial electron transfer to dioxygen or other electron acceptors,3,11 which has been supported by electrochemical and time-resolved spectroscopic investigations.12-14 However, the reported Pt effects have not been always positive15-18 and the role of Pt deposits in photocatalytic reaction mechanisms seems to be far more complex than being a simple electron sink. For example, the photocatalytic oxidation of CO was highly enhanced on Pt/ TiO2,10a,19 but the enhancement effect was not observed at a lower temperature (105 K) at which most thermal effects should be quenched.20 The thermal catalytic role of Pt is also important in the overall behavior of Pt/TiO2.21-23 On the other hand, the reported Pt effects in photocatalytic degradation (PCD) of a specific substrate are often contradictory. For example, Zhao et al.24 reported that the PCD of Rhodamine-B (RB) dye on Pt/TiO2 was 3 times as fast as its PCD on naked TiO2, whereas Muradov25 observed that the platinization of TiO2 did not change the PCD rate of RB significantly. Enhanced PCDs of phenolic compounds on Pt/TiO2 have not been observed in many studies.11,17,26 Reported experimental results about the PCD of trichloroethylene (TCE) on Pt/TiO2 are * Corresponding author. E-mail:
[email protected]. Fax: +82-54279-8299.
contradictory as well. Chen et al.18 reported that the platinization of TiO2 drastically reduced the PCD rate of TCE, and Driessen et al.27 observed a similar phenomenon. On the contrary, Crittenden et al.28 reported that such a significant retardation in PCD of TCE was not observed with Pt/TiO2. This study aims to understand what specific parameters control the photocatalytic reaction when Pt/TiO2 is used and why reported Pt effects are inconsistent for some substrates. We investigated the platinization effects in the TiO2-mediated PCD of several chlorinated organic compounds that include trichloroethylene, perchloroethylene, dichloroacetate, trichloroacetate, chloroform, and 4-chlorophenol. The Pt effects in PCD reactions were highly substrate-specific and were sensitively influenced by the condition of Pt deposition. The oxidation state of deposited Pt species was found to be the most important factor in determining the PCD activity. Experimental Section Chemicals and Materials. Chemicals that were used as received in this study include trichloroethylene (TCE, Junsei), perchloroethylene (PCE, 98%, Kanto), chloroform (CF, >99%, Matsunoen), 4-chlorophenol (4-CP, extra pure, Junsei), sodium dichloroacetate (DCA, >98% Aldrich), trichloroacetate (TCA, Aldrich), chloroplatinic acid (H2PtCl6, 99.995%, Aldrich), phosphotungstic acid (HNa2PW12O40, or polyoxometalate, Riedelde Haen). Water used was ultrapure (18 MΩ cm) and prepared by a Barnstead purification system. Titanium dioxide (Degussa P25), a mixture of 80% anatase and 20% rutile, was used as a base photocatalyst. Other chemicals used were of the highest purity available. Preparation of Platinized TiO2. Platinization of TiO2 was carried out using a photodeposition method. An air-equilibrated aqueous suspension of TiO2 (0.5 g/L) was illuminated with UV
10.1021/jp044425+ CCC: $30.25 © 2005 American Chemical Society Published on Web 03/17/2005
7400 J. Phys. Chem. B, Vol. 109, No. 15, 2005 in the presence of chloroplatinic acid (Pt precursor) and methanol (electron donor). To investigate the effect of deposited Pt speciation, the platinization was carried out under three different conditions: (1) an electron-donor-rich (ED-rich) condition with 1 M methanol and 0.1 mM chloroplatinic acid (illuminated with a 200-W mercury lamp) at pH 3, (2) an electron-donor-deficient (ED-deficient) condition with 0.1 M methanol and 1 mM chloroplatinic acid (illuminated with a 300-W Xe arc lamp) at pH 3, and (3) an electron-donor-free (ED-free) condition with 0.1 mM chloroplatinic acid at pH 10 (illuminated with a 200-W mercury lamp). Note that the ratio of [methanol]/[H2PtCl6] in the ED-rich condition is higher by 100 times than that in the ED-deficient condition. After irradiation, the Pt/TiO2 powder was collected by filtration and washed with distilled water. When needed, Pt/TiO2 that was prepared by the photodeposition was further treated by reducing agents such as NaBH4 and reduced polyoxometalate (described later). The concentration of unused chloroplatinic acid remaining in the filtrate solution after the photodeposition was determined by inductively coupled plasma atomic emission spectroscopy (ICP-AES, Thermo Jarrell Ash Corp. IRIS/AP) in order to quantify the amount of deposited Pt. Transmission electron micrographic images of Pt particles loaded on TiO2 were obtained using a JEOL 2100F high-resolution transmission electron microscope (HRTEM). The oxidation states of Pt deposited on TiO2 catalysts were analyzed by X-ray photoelectron spectroscopy (XPS, Kratos XSAM 800pci) using Mg KR lines (1253.6 eV) as an excitation source. The spectra were taken for each sample after Ar+ (3 keV) sputter cleaning. Surface charging was minimized by spraying low energy electrons over the sample using a neutralizer gun. Binding energy spectra were recorded in the region of C 1s, Ti 2p, O 1s, and Pt 4f. The binding energies of all peaks were referenced to the Ti 2p line (458.8 eV) in TiO2 and checked against the C 1s line (284.6 eV) originating from surface impurity carbons. Photolysis and Analysis. All naked or platinized TiO2 suspensions were prepared at a concentration of 0.5 g/L and were dispersed by simultaneous sonication and shaking for 30 s in an ultrasonic cleaning bath. Saturated stock solutions of TCE (8.4 mM), PCE (2.45 mM), and CF (64 mM) were prepared in distilled water containing excess substrates, and desired concentrations were made by dilution. As for 4-CP and DCA, an aliquot of stock solution (10 mM) was added to the TiO2 suspension to make a desired concentration. All photolysis experiments in this work were carried out at initial pH of 4.0. The photolyses were performed with a 300-W Xe-arc lamp (Oriel). Light passed through a 10-cm IR water filter and a UV cutoff filter (λ > 300 nm). The filtered light was focused onto a 60-mL Pyrex reactor with a quartz window. The photocatalytic reactor was filled with minimized headspace, sealed with a rubber septum, and stirred magnetically. Sample aliquots of 1 mL were withdrawn from the illuminated reactor with a 1-mL syringe, filtered through a 0.45-µm poly(tetrafluoroethylene) (PTFE) filter (Millipore), and injected into a 2-mL glass vial. Quantitative analysis of chloride ions generated as a result of PCD of chlorinated substrates was performed by using an ion chromatograph (IC, Dionex DX-120). The IC system was equipped with a Dionex IonPac AS-14 for anions and a conductivity detector. The analysis of 4-CP was done by using a high performance liquid chromatograph (HPLC, Agilent 1100) equipped with a C-18 column (Agilent Zorbax 300SB) and a diode-array detector (DAD). The rates of chloride generation
Lee and Choi or substrate degradation were linear during the initial 30 min. The PCD rates were expressed in terms of the initial rate. Photoelectrochemical Measurements. The photoelectrochemical characteristics of naked TiO2, Ptox/TiO2, and Pt0/TiO2 electrodes were compared. TiO2 electrodes were prepared as described elsewhere.29 Degussa P25 TiO2 (0.2 g) was added to 2 g of 50 wt % aqueous Carbowax (poly(ethylene glycol)) solution as a binder and thoroughly mixed. The resulting paste was cast on an indium tin oxide (ITO) glass (1 × 1 cm2, Samsung) with two tracks of one layer of Scotch Magic Tape and then left at room temperature for 30 min. The dried film was heated at 400 °C for 30 min to burn off any organics on the TiO2 film surface. For Pt deposition onto a TiO2 electrode, the electrode was immersed in an aqueous solution of 1 mM H2PtCl6 and 0.1 M methanol (ED-deficient condition) and irradiated with a 200-W Hg lamp for 30 min. The XPS spectra of Pt (4f band) deposited on the TiO2 electrode showed that the main Pt species was PtO2. The resulting electrode was Ptox/ TiO2. The Pt0/TiO2 electrode could be prepared by chemically reducing PtO2 on the TiO2 electrode using NaBH4 reagent. The oxidation state of Pt0 was confirmed by the XPS analysis. The photoelectrochemical reactor was cylindrical and had a working electrode (TiO2, Ptox/TiO2, or Pt0/TiO2), a reference saturated calomel electrode (SCE), and a graphite rod as a counter electrode. Photocurrents were collected by applying a potential (+0.5 V vs SCE) to the working electrode using a potentiostat (EG&G 263A2) connected to a computer. The photocurrent generation in the presence of chlorinated organic substrates was compared among different TiO2 electrodes (TiO2, Ptox/TiO2, and Pt0/TiO2). A 300-W Xe arc lamp (Oriel) was used as a UV light source as in PCD experiments. Results and Discussion Substrate-Specific Pt Effect in PCD Reactions. The PCD rates of TCE, PCE, DCA, 4-CP, and CF in air-equilibrated aqueous suspensions are compared in Figure 1 when a series of Pt/TiO2 samples prepared with varying Pt photodeposition time (1, 2, 5, 30, and 60 min) were used as a photocatalyst. The platinization was carried out in the ED-rich condition. As for DCA degradation, a volcano-shaped activity curve, which has been frequently observed in other studies,11,24,26,30 was obtained. In general, the PCD rate increases with Pt loading because of the efficient charge separation at the interface of Pt and TiO2 but too much Pt loading decreases the photoreactivity mainly because active sites on TiO2 are covered by Pt. However, the PCD dependence of TCE and PCE on the Pt loading is drastically different from the general case. TiO2 that was photoplatinized for 1, 2, and 5 min had detrimental effects on the PCD rates of TCE and PCE, while the 30-min photodeposition of Pt recovered the photoactivity to a level similar to that of naked TiO2. Further photodeposition for 60 min decreased the TCE and PCE degradation rate as in the case of DCA. The effect of Pt deposition on the PCD of 4-CP was similar to the case of DCA, whereas the CF case did not show a sign of activity decrease up to 60 min of photodeposition (Figure 1b). This result clearly demonstrates that the Pt loading effects in TiO2 photocatalytic reactions depend on not only the Pt loading but also the kind of substrates, and hence cannot be generalized. This also provides a reasonable explanation of why the Pt loading effects reported in the literature are often inconsistent even for the same substrate (e.g., TCE). The term “platinized TiO2” or simply “Pt/TiO2” may represent very different photocatalysts depending on the preparation conditions. The characteristics of a “Pt/TiO2” catalyst should depend on various
Photocatalytic Reactivity of Platinized TiO2
Figure 1. Effects of Pt photodeposition time on the PCD rates of (a) TCE, PCE, DCA and (b) 4-CP, CF. The experimental conditions were [Pt/TiO2] ) 0.5 g/L; [TCE]0 ) 0.4 mM; [PCE]0 ) 0.25 mM; [DCA]0 ) 1 mM; [4-CP]0 ) 0.25 mM; [CF]0 ) 1 mM; pH ) 4.0; air-equilibrated [note that the abscissa scale (photodeposition time) is not linear]. The photoplatinization was carried out under the ED-rich condition.
parameters such as the property of substrate TiO2, the loaded Pt amount, the oxidation state of Pt, the particle size and dispersity of Pt, etc. Effects of Pt Loading and Size. The properties of deposited Pt change with the photodeposition time. The amount of Pt deposited on TiO2 was determined by ICP-AES analysis. The Pt loading was not proportional to the photodeposition time: the loaded Pt was 2.2 and 2.5 wt % for 1- and 2-min photodeposition and remained constant around 3.3 wt % (corresponding to quantitative deposition of all platinic acids) for 5-min photodeposition and beyond. The XPS analysis of surface atomic compositions also showed a minor effect of photodeposition time on the Pt loading: the surface Pt concentration increased only from 1.0 to 1.4 at. % between the photodeposition times of 1 and 60 min. The minor variation of Pt loading on TiO2 in this range of photodeposition time does not seem to be responsible for the sensitive response of the photoreactivity to the photodeposition time (Figure 1). In particular, the drastic retardation in the PCD rates of TCE and PCE observed with Pt/TiO2 (1-5 min of photodeposition) cannot be explained by the shielding effect that Pt coverage blocks reactive surface sites.24,30 On the other hand, the size of Pt particles was dependent on the photodeposition time and the electron donor concentration.
J. Phys. Chem. B, Vol. 109, No. 15, 2005 7401 Figure 2 compares the TEM images of Pt/TiO2 catalysts that were prepared in the ED-rich (a,b) and ED-deficient conditions (c,d), respectively. The TEM image of Pt/TiO2 (ED-free, image not shown) was similar to that of Pt/TiO2 (ED-deficient). For Pt/TiO2 prepared in the ED-rich condition, Pt particles on Pt/ TiO2 (1 min) are too small to be clearly seen but those on Pt/ TiO2 (30 min) are well defined with the size of 2-10 nm. The metal particles with different size may differ in their activities for the interfacial electron transfer and catalysis on TiO2. For example, Subramanian et al.31 observed that the electron transfer to C60 on Au/TiO2 increased when the size of Au nanoparticles decreased. However, in the present case of Pt/TiO2, the size of Pt particles does not seem to be a main parameter in determining the photocatalytic activity since smaller Pt particles (Figure 2a) were less active than larger ones (Figure 2b) for the PCD of all substrates tested. As for Pt/TiO2 prepared in the ED-deficient condition, the size of Pt particles does not change with the photodeposition time (compare Figure 2c with Figure 2d) and are smaller than those made in the ED-rich condition (compare Figure 2b with Figure 2d). As discussed later, Pt/TiO2 (EDdeficient) is less active than Pt/TiO2 (ED-rich), which also shows that smaller Pt particles on TiO2 are not more active than larger ones for PCD reactions. The size of Pt particles should not be the main deciding factor in controlling the PCD activity. Effect of Pt Oxidation State. Photodeposited Pt on a TiO2 surface reportedly exists in various oxidation states such as PtIV, PtII, and Pt0 species.32 The main effect of photodeposition time might be related to the change in the oxidation state of deposited Pt species, and the Pt oxidation state can be controlled by changing the preparation condition. PtCl62- in aqueous solution can be hydrolyzed to form Pt(OH)xCl6-x2- (x ) 1-5), which is pHdependent.33 Each Pt(OH)xCl6-x2- species should differ in its reactivity for conduction band (CB) electrons and valence band (VB) holes. For example, it has been reported that PtO2 and PtO (or Pt(OH)2) were deposited in alkaline conditions through the reaction of hydrolyzed Pt complexes with VB holes (e.g., Pt(OH)4Cl22- + 4hvb+ f PtO2 + 4H+ + O2 + 2Cl-) whereas Pt0 was mainly deposited at acidic pH.33,34 The degree of the hydrolysis of PtCl62- increases with pH, and the formation of Pt oxides seems to be favored at basic conditions.34 Therefore, Pt oxides were deposited in the ED-free condition at pH 10 in this study. On the contrary, the formation of Pt0 should be favored in acidic conditions and in the presence of electron donors that prevent the Pt complexes from reacting with VB holes. We prepared Ptox/TiO2 and Pt0/TiO2 under different photodeposition conditions: the former in the ED-free condition (pH 10) and the latter in the ED-rich condition (pH 3). The XPS spectra in Figure 3a show that Ptox/TiO2 and Pt0/TiO2 have PtIV/ PtII and Pt0 as main species, respectively. Although both Pt/ TiO2 samples had a similar Pt content, their photocatalytic activities were very different as shown in Figure 3b,c. Ptox/TiO2 is particularly inactive for TCE degradation, and Pt0/TiO2 is generally more active than Ptox/TiO2 except for 4-CP whose degradation appears to be little affected by the Pt oxidation state. Pt0/TiO2 was more active than Ptox/TiO2 not only for the oxidative degradation of chlorinated organic compounds (TCE, DCA, CF) but also for the reductive degradation of TCA.22 Incidentally, we also carried out the Pt photodeposition in the deaerated condition. The Pt/TiO2 sample prepared in the deaerated suspension (ED-rich condition) exhibited a higher fraction of Pt0 and hence a higher activity for TCE degradation than that prepared in the presence of oxygen (data not shown). Because O2 should compete with chloroplatinic acids for CB
7402 J. Phys. Chem. B, Vol. 109, No. 15, 2005
Lee and Choi
Figure 2. TEM images of Pt/TiO2 catalysts prepared by (a) 1-min and (b) 30-min Pt photodeposition under the ED-rich condition and (c) 5-min and (d) 120-min Pt photodeposition under the ED-deficient condition.
electrons, the photoreductive conversion from PtIV to Pt0 is more favored in the absence of O2. PtII/PtIV species on TiO2 may act as a recombination center through the consecutive reduction/oxidation cycles consuming considerable amounts of photogenerated electrons and holes (reaction 1),35 which explains why Ptox/TiO2 is less active than Pt0/TiO2. e-
e-
e-
e-
h
h
h
h
PtIV y\ z PtIII y\ z PtII y\ z PtI y\ z Pt0 + + + +
(1)
In addition, Pt0 metal serves as a thermal catalyst and changes the photocatalytic reaction pathways and products.21-23,36,37 We recently demonstrated that aqueous ammonia can be converted directly into dinitrogen on illuminated Pt/TiO221 and that anoxic PCD of TCA is highly enhanced on Pt/TiO2 since the presence of Pt on TiO2 provides an alternative reaction path.22 In a similar context, dimethylamine can transform into trimethylamine on Pt/TiO2 in deaerated solution whereas only demethylated products were generated with pure TiO2.23 To investigate the effect of photodeposition time on the oxidation state of Pt, the XPS spectra of Pt/TiO2 catalysts that were prepared with varying photodeposition time are compared in Figure 4. The oxidation states of Pt deposited in (a) the EDrich condition (the same condition used in Figure 1) and (b)
the ED-deficient condition show very different dependence on the photodeposition time. As for the Pt/TiO2 prepared in the ED-rich condition (Figure 4a), PtII was the major species in the early photodeposition period (1-2 min) but Pt0 was dominant after longer photodeposition (30-60 min). This confirms that the photoactivity variation of Pt/TiO2 in TCE degradation (Figure 1a) should be ascribed to the fact that the Pt oxidation state changes from PtII to Pt0 with the photodeposition time. That is, the early photodeposition produces Ptox/TiO2 that is particularly inactive for TCE degradation but further photodeposition recovers the photoactivity since Ptox/TiO2 transforms into Pt0/TiO2. On the other hand, Pt photodeposition under the ED-deficient condition (Figure 4b) did not show such a change in the oxidation state as the photodeposition time increases. The formation of Pt0 species was insignificant up to 120 min of photodeposition, and PtII was the dominant species throughout the photodeposition. Therefore, the TEM images of Figure 2a,c,d represent Ptox/TiO2 while Figure 2b shows the image of Pt0/ TiO2. It has been previously reported that platinized TiO2 having PtO (or Pt(OH)2) or PtO2 as the main Pt species could be prepared through photodeposition in the absence of an electron donor.38 The successive reduction from PtIV to Pt0 should be driven by not only CB electrons but also methanol radicals (•CH2OH) that are much stronger reductants than the former
Photocatalytic Reactivity of Platinized TiO2
J. Phys. Chem. B, Vol. 109, No. 15, 2005 7403
Figure 4. XPS spectra of Pt (4f band) on TiO2 that was loaded with varying the time of photodeposition (PD), which was carried out under two different conditions: (a) 0.1 mM chloroplatinic acid and 1 M methanol (ED-rich) and (b) 1 mM chloroplatinic acid and 0.1 M methanol (ED-deficient).
Figure 3. (a) XPS spectra of Pt (4f band) on Ptox/TiO2 and Pt0/TiO2. Ptox/TiO2 was prepared with 30-min photodeposition in the ED-free condition, and Pt0/TiO2 with 60-min photodeposition in the ED-rich condition. Comparison of the PCD of (b) TCE, DCA and (c) CF, TCA, 4-CP with naked TiO2, Ptox/TiO2, and Pt0/TiO2 ([TCE]0 ) 0.4 mM; [DCA]0 ) 0.5 mM; [CF]0 ) 1 mM; [TCA]0 ) 0.5 mM; [4-CP]0 ) 0.25 mM; pHi ) 4.0). All suspensions were air-equilibrated except for the case of TCA that was carried out in the deaerated condition.
(E° ) -0.74 VNHE for the •CH2OH/HCHO couple).6f It seems that the CB electrons alone in the ED-free condition or the methanol radicals in the ED-deficient condition are not sufficient to reduce PtIV fully to Pt0. Figure 5 shows the photoreactivities of Pt/TiO2 that was prepared in the ED-deficient condition (i.e., Pt/TiO2 of Figure 4b) for the PCD of DCA and TCE. Figure 5 can be directly compared with Figure 1a that shows the photoreactivities of Pt/TiO2 prepared in the ED-rich condition. The two Pt/TiO2
catalysts are very different in their photoactivities: the PCD rate of TCE with Pt/TiO2 in Figure 5 was not recovered at all at longer photodeposition time, whereas that in Figure 1a rebounded after 30-min photodeposition. The most plausible explanation for the different behaviors between the two Pt/TiO2 samples is the different oxidation state of deposited Pt species. The photoreactivity recovery (in TCE degradation) shown in Figure 1a coincides with the appearance of Pt0 species (Figure 4a); the failure in recovering the reactivity at longer photodeposition time (Figure 5) is consistent with the absence of Pt0 species on TiO2 (Figure 4b). On the other hand, the PCD of DCA seems to be much less affected by the Pt oxidation state. The oxidation states of Pt on TiO2 might be further changed through post-photodeposition processes. For example, if Pt oxidation states can change gradually during photocatalytic reactions, the Pt speciation on fresh and used Pt/TiO2 catalysts could be different. In such a case, the photocatalytic reactivity of a Pt/TiO2 catalyst may change with repeated uses, which is exactly what Figure 6 shows. The photocatalytic reactivity of Pt/TiO2 for TCE degradation gradually increased with repeated uses, whereas that for DCA degradation was little affected. The XPS spectra in Figure 6c show that a fraction of Pt species were reduced to Pt0 after use. This reconfirms that the PCD of TCE strongly depends on the oxidation state of Pt on TiO2.
7404 J. Phys. Chem. B, Vol. 109, No. 15, 2005
Lee and Choi
Figure 5. The initial rate of chloride production from the PCD of TCE and DCA with Pt/TiO2 as a function of Pt photodeposition time ([TCE]0 ) 0.4 mM; [DCA]0 ) 1 mM; pH ) 4.0). The photoplatinization was carried out in the ED-deficient condition.
As an alternative method to change the oxidation state of Pt photodeposited on TiO2, Pt/TiO2 was additionally reacted with a reducing agent in the dark after the completion of photodeposition. A polyoxometalate (POM), PW12O403-, was used as the reducing agent: reduced POMs (POM-) are able to reduce metal ions (e.g., Ag+, Pd2+, AuCl4-, and PtCl62-) into metal nanoparticles.39,40 POM (PW12O403-, 10 mM) was rapidly reduced to POM- (PW12O404-) in UV-illuminated TiO2 suspension (0.5 g/L, deaerated) that contained methanol (6.25 M) as an electron donor (reaction 2).
PW12O403- + ecb- f PW12O404-
(2)
After 2 h of UV illumination, the formation of POM- was confirmed by the deep blue coloration of the suspension. The colored suspension containing POM- was filtered through a syringe filter (0.45-µm PTFE filter) to remove TiO2 particles and then immediately added into a deaerated Pt/TiO2 suspension (0.25 g/L). The mixed suspension was left for 1 h in the dark to allow POM- to reduce Ptox species on TiO2 (reaction 3).
nPOM- + Ptn+/TiO2 f nPOM + Pt0/TiO2
(3)
The XPS spectra in Figure 7a clearly show that the above POM treatment based on reaction 3 indeed reduced the surface Ptox species into Pt0. As a result, the PCD rate of TCE was markedly enhanced with the POM-treated Pt/TiO2 (Figure 7b). We have shown that the oxidation state of Pt deposited on TiO2 can be changed in several ways. The conditions of photodeposition, reusing Pt/TiO2, and chemical treatment following the photodeposition all influenced the oxidation state of Pt. All the cases consistently showed that the presence of oxidized Pt species on TiO2 strongly retarded the PCD of TCE but reducing Ptox to Pt0 rapidly recovered the photoactivity of Pt/TiO2 for TCE degradation. However, DCA degradation was much less influenced by the reduction of oxidized Pt species on TiO2. Incidentally, the present Ptox/TiO2 did not exhibit any visible light activity for 4-CP degradation whereas Zang et al.41 reported that TiO2 modified with chlorides of PtIV catalyzed the degradation of 4-CP under visible light.
Figure 6. PCD of (a) TCE and (b) DCA with fresh and reused Pt/ TiO2 catalyst ([TCE]0 ) 0.4 mM; [DCA]0 ) 1 mM; pH ) 4.0). (c) XPS spectra of Pt (4f band) on fresh and reused Pt/TiO2. The fresh Pt/TiO2 was prepared with 2-min photodeposition in the ED-rich condition.
Pt Effects on Photocurrent and PCD Mechanism. To investigate the effects of platinization on the electron transfer at the TiO2/water interface, the photocurrents (Iph) generated with TiO2, Ptox/TiO2, and Pt0/TiO2 electrodes in the presence of DCA or TCE are compared in Figure 8. Once CB electrons are photogenerated by the band gap excitation of the Pt/TiO2
Photocatalytic Reactivity of Platinized TiO2
Figure 7. (a) XPS spectra of Pt (4f band) that was deposited on TiO2 in two different ways: one following a normal photodeposition procedure (2-min deposition in the ED-rich condition), Pt/TiO2 (PD), and the other combining the photodeposition and the subsequent chemical reduction by reduced POM, Pt/TiO2 (PD + POM-). (b) Comparison of the photocatalytic reactivity for TCE degradation between Pt/TiO2 (PD) and Pt/TiO2 (PD + POM-) ([TCE]0 ) 0.4 mM; pH ) 4.0).
electrode, they are recombined with VB holes directly or indirectly, trapped in defect sites or Pt deposits, transferred to electron acceptors (e.g., O2) in the solution, or extracted into the external circuit to generate Iph. The photocurrent profiles shown in Figure 8 indicate that the Pt deposits on TiO2 indeed trap CB electrons to decrease Iph. At the steady state, the photocurrent decreased in the order of TiO2 > Pt0/TiO2 > Ptox/ TiO2 for both cases of DCA and TCE. Similar trends were also observed with PCE, 4-CP, or pure water. The fact that Iph was lower with Ptox/TiO2 than Pt0/TiO2 implies either that Ptox species trap more CB electrons or that Ptox species serve as a recombination center. Because Ptox/TiO2 was less active than Pt0/TiO2 for all PCD reactions tested in this study, the lower photocurrents obtained with the Ptox/TiO2 electrode should be ascribed to the faster recombination in Ptox/TiO2. However, this does not explain why Ptox/TiO2 is particularly inactive for TCE (or PCE) degradation. The substrate-specific PCD activities of Pt/TiO2 should be related to the intrinsic difference in the degradation mechanisms among various substrates. Here we propose a hypothesis that TCE (or PCE) has particularly strong chemical affinity with the Ptox species on which the charge-pair recombination is chemically mediated through the redox cycling of TCE (or PCE). The specific chemical interaction between chlorinated ethenes and Ptox needs to be verified in further studies. Scheme
J. Phys. Chem. B, Vol. 109, No. 15, 2005 7405
Figure 8. Time profiles of photocurrent, Iph, obtained with pure TiO2, Ptox/TiO2, and Pt0/TiO2 electrodes in the presence of (a) DCA ([DCA]0 ) 1 mM) and (b) TCE ([TCE]0 ) 1 mM) under the air-equilibrated condition (pH 4). The inset shows the XPS spectra of the Pt 4f band of Pt0/TiO2 and Ptox/TiO2 electrodes.
SCHEME 1: Proposed Photoinduced Electron Transfer Paths on Ptox/TiO2 and Pt0/TiO2 in the Presence of TCEa
a The numbers indicate major electronic pathways: (1) band gap excitation; (2) electron trapping in the Pt0 phase; (3) electron trapping in the Ptox phase; (4) trapped electron transfer to O2; (5) Ptox-mediated recombination; (6) trapped electron transfer to TCE; (7) TCE-mediated recombination; (8) reductively initiated degradation of TCE.
1 illustrates the situation. Although both Pt0 and Ptox deposits can trap CB electrons that are subsequently transferred to electron acceptors such as O2, Ptox also serves as a recombination
7406 J. Phys. Chem. B, Vol. 109, No. 15, 2005 center on the contrary. This seems to be why Ptox/TiO2 is more active than pure TiO2 but less active than Pt0/TiO2 in general. When TCE is adsorbed on the Ptox surface, it provides an additional recombination path in which TCE reacts successively with trapped electrons and holes (step 6 + 7 in Scheme 1). This chemically mediated recombination path in the presence of TCE may explain the observation that Iph generated in the presence of TCE was consistently lower than Iph in the presence of DCA (Figure 8). Once TCE•- radical anions are formed after accepting CB electrons, a small fraction of them may follow a reductive degradation path (step 8 in Scheme 1). TCE could be degraded in the absence of dissolved oxygen through this reductive path. Although most PCD reactions of organic compounds are almost completely inhibited in the absence of dissolved O2, anoxic PCD pathways exist in special cases. For example, such anoxic mechanism can be highly enhanced by the surface platinization of TiO2.22,23 The reductive degradation of TCE in the anoxic condition seems to accelerate on Pt0/TiO2. In N2-saturated suspensions, the PCD rates of TCE were markedly enhanced in the presence of Pt0 deposits on TiO2 (5.4 µM/min with Pt0/ TiO2 versus 1.2 µM/min with TiO2) whereas those in the airequilibrated suspensions were little affected by the surface platinization (13.3 µM/min with Pt0/TiO2 versus 13.5 µM/min with TiO2). This indicates that the anoxic degradation is initiated by the CB electron transfer to TCE (step 6 in Scheme 1), which occurs preferentially on the platinized surface. Conclusions The effects of platinization of TiO2 on the photocatalytic reactivity should depend on many experimental factors that include not only the properties of Pt deposits (e.g., loaded amount, oxidation state, size, etc.) but also the interaction between substrates and Pt deposits. Although the interplay of all these parameters cannot be fully addressed in a single study, this work clearly demonstrated that the platinization effects are substrate-specific and hard to generalize. The case of TCE and PCE is a prominent example. Many studies on platinized TiO2 have been reported in the literature, but they employed various methods of platinization to make “Pt/TiO2” photocatalysts that should be quite different in many aspects. In addition, their photocatalytic reactivities were tested for a wide variety of substrates, which makes the reliable comparison of published results on the activity of Pt/TiO2 very difficult. In particular, it should be noted that the oxidation state of Pt species is the most important parameter in determining the PCD activity of Pt/TiO2 and needs to be carefully controlled to obtain the desired activity. Acknowledgment. This work was supported by KOSEF through the Center for Integrated Molecular Systems (CIMS) and partly by the Brain Korea 21 project. References and Notes (1) Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W. Chem. ReV. 1995, 95, 69. (2) Photocatalytic Purification and Treatment of Water and Air; Ollis, D. F., Al-Ekabi, H., Eds.; Elsevier: Amsterdam, 1993.
Lee and Choi (3) Photocatalysis-Fundamentals and Applications; Serpone, N., Pelizzetti, E., Eds.; Wiley-Interscience: New York, 1989. (4) Ollis, D. F. EnViron. Sci. Technol. 1985, 19, 480. (5) Matthews, R. W. J. Phys. Chem. 1987, 91, 3328. (6) (a) Lee, M. C.; Choi, W. J. Phys. Chem. B 2002, 106, 11818. (b) Lee, H.; Choi, W. EnViron. Sci. Technol. 2002, 36, 3872. (c) Choi, W.; Hong, S. J.; Chang, Y. S.; Cho, Y. EnViron. Sci. Technol. 2000, 34, 4810. (d) Kim, S.; Choi, W. EnViron. Sci. Technol. 2002, 36, 2019. (e) Cho, S.; Choi, W. J. Photochem. Photobiol. A 2001, 143, 221. (f) Choi, W.; Hoffmann, M. R. EnViron. Sci. Technol. 1995, 29, 1646. (7) Kraeutler, B.; Bard, A. J. J. Am. Chem. Soc. 1978, 100, 4317. (8) Baba, R.; Nakabayashi, S.; Fujishima, A.; Honda, K. J. Phys. Chem. 1985, 89, 1902. (9) Ranjit, K. T.; Viswanathan, B. J. Photochem. Photobiol. A 1997, 108, 73. (10) (a) Hwang, S.; Lee, M. C.; Choi, W. Appl. Catal. B 2003, 46, 49. (b) Choi, W.; Lee, J.; Kim, S.; Hwang, S.; Lee, M. C.; Lee, T. K. J. Ind. Eng. Chem. 2003, 9, 96. (11) Hufschmidt, D.; Bahnemann, D.; Testa, J. J.; Emilio, C. A.; Litter, M. I. J. Photochem. Photobiol. A 2002, 148, 223. (12) Yamakata, A.; Ishibashi, T.; Onishi, H. J. Phys. Chem. B 2001, 105, 7258. (13) Choi, Y.-K.; SeO, S.-S.; Chjo, K.-H.; Choi, Q-W.; Park, S.-M. J. Electrochem. Soc. 1992, 139, 1803. (14) Furube, A.; Asahi, T.; Masuhara, H.; Yamashita, H.; Anpo, M. Chem. Phys. Lett. 2002, 336, 424. (15) Klare, M.; Scheen, J.; Vogelsang, K.; Jacobs, H.; Broekaert, J. A. C. Chemosphere 2000, 41, 353. (16) Einaga, H.; Futamura, S.; Ibusuki, T. EnViron. Sci. Technol. 2001, 35, 1880. (17) Sun, B.; Vorontsov, V.; Smirniotis, P. G. Langmuir 2003, 19, 3151. (18) Chen, J.; Ollis, D. F.; Rulkens, W. H.; Bruning, H. Water Res. 1999, 33, 661. (19) Einaga, H.; Harada, M.; Futamura, S.; Ibusuki, T. J. Phys. Chem. B 2003, 107, 9290. (20) Linsebigler, A.; Rusu, C.; Yates, J. T., Jr. J. Am. Chem. Soc. 1996, 118, 5284. (21) Lee, J.; Park, H.; Choi, W. EnViron. Sci. Technol. 2002, 36, 5462. (22) Kim, S.; Choi, W. J. Phys. Chem. B 2002, 106, 13311. (23) Lee, J.; Choi, W. EnViron. Sci. Technol. 2004, 38, 4026. (24) Zhao, W.; Chen, C.; Li, X.; Zhao, J.; Hidaka, H.; Serpone, N. J. Phys. Chem. B 2002, 106, 5022. (25) Muradov, N. Z. Solar Energy 1994, 52, 283. (26) Trillas, M.; Peral, J.; Domenech, X. Appl. Catal. B 1995, 5, 377. (27) Driessen, M. D.; Grassian, V. H. J. Phys. Chem. B 1998, 102, 1418. (28) Crittenden, J. C.; Liu, J.; Hand, D. W.; Perram, D. L. Water Res. 1997, 31, 429. (29) Mills, A.; Elliot, N.; Hill, G.; Fallis, D.; Durrant, J. R.; Wills, R. L. Photochem. Photobiol. Sci. 2003, 2, 591. (30) Bae, E.; Choi, W. EnViron. Sci. Technol. 2003, 37, 147. (31) Subramanian, V.; Wolf, E. E.; Kamat, P. V. J. Am. Chem. Soc. 2004, 126, 4943. (32) Sungbom, C.; Kawai, M.; Tanaka, K. Bull. Chem. Soc. Jpn. 1984, 58, 871. (33) Jin, Z.; Chen, Z.; Li, Q.; Xi, C.; Zheng, X. J. Photochem. Photobiol. A 1994, 81, 177. (34) Zhang, F.; Chen, J.; Zhang, X.; Gao, W.; Jin, R.; Guan, N.; Li, Y. Langmuir 2004, 20, 9329. (35) Subramanian, V.; Wolf, E. E.; Kamat, P. V. Langmuir 2003, 19, 469. (36) Ohtani, B.; Iwai, K.; Nishimoto, S.; Sato, S. J. Phys. Chem. B 1997, 101, 3349. (37) Bahnemann, D. W.; Mo¨nig, J.; Chapman, R. J. Phys. Chem. 1987, 91, 3782. (38) Sano, T.; Negishi, N.; Uchino, K.; Tanaka, J.; Matsuzawa, S.; Takeuchi, K. J. Photochem. Photobiol. A 2003, 160, 93. (39) Troupis, A.; Hiskia, A.; Papaconstantinou, E. Angew. Chem., Int. Ed. 2002, 41, 1911. (40) Mandal, S.; Selvakannan, PR.; Pasricha, R.; Sastry, M. J. Am. Chem. Soc. 2003, 125, 8440. (41) Zang, L.; Macyk, W.; Lange, C.; Maier, W. F.; Antonius, C.; Meissner, D.; Kisch, H. Chem.sEur. J. 2000, 6, 379.
